Introduction
The goal of therapy for patients with rheumatoid arthritis (RA) is to achieve sustained remission or low disease activity [
1,
2]. For patients who fail to achieve these goals with first-line therapy, typically conventional synthetic disease-modifying anti-rheumatic drugs (csDMARDs), treatment options include biologic DMARDs (bDMARDs) such as tumor necrosis factor (TNF) inhibitors, or Janus kinase (JAK) inhibitors [
2]. Upadacitinib is a JAK inhibitor that has demonstrated efficacy in RA with an acceptable safety profile; after 24/26 weeks of upadacitinib 15-mg treatment, approximately one-quarter of patients with inadequate response to methotrexate (MTX)/csDMARDs achieved clinical remission (Clinical Disease Activity Index [CDAI] ≤ 2.8) [
3‐
10]. While the management of RA should include regular monitoring of disease activity in accordance with treat-to-target principles, consideration of patient safety is equally important, with monitoring of adverse events (AEs) and commonly tested laboratory parameters.
JAK inhibitors are orally administered small molecule drugs that have been shown to be effective for the treatment of moderate-to-severe RA in multiple randomized controlled trials and are recommended treatment options for patients who fail to respond to csDMARDs and/or bDMARDs [
2,
11,
12]. JAK inhibitors have been associated with changes in laboratory parameters including decreases in lymphocytes, neutrophils, and hemoglobin, increases in the liver enzymes alanine aminotransferase (ALT) and aspartate aminotransferase (AST), increases in serum creatinine and creatine phosphokinase (CPK), and increases in blood cholesterol [
12‐
14]. As a result, it is recommended that patients receiving JAK inhibitors undergo routine laboratory monitoring [
13,
15‐
18].
The JAK inhibitor upadacitinib has been engineered for increased selectivity for JAK1 over JAK2, JAK3, and tyrosine kinase 2 at therapeutic concentrations [
18,
19]. Upadacitinib has been evaluated across a broad spectrum of patients in the global phase 3 SELECT program, which included six trials with long-term open-label extensions [
18,
20,
21]. The SELECT program included patients naïve to MTX, patients with an inadequate response to MTX/csDMARDs, and patients with an inadequate response to bDMARDs. It assessed upadacitinib both as monotherapy and in combination with MTX [
3‐
8].
Consistent with the known safety profile of JAK inhibitors, changes in laboratory parameters have been observed in upadacitinib clinical trials [
9,
20]. In this descriptive integrated analysis, we assessed selected long-term laboratory profiles associated with exposure to upadacitinib and active controls (adalimumab and MTX) in patients with RA treated in the SELECT program, with up to 6.5 years of exposure.
Methods
Patients and Studies
Data from six randomized phase 3 trials and their open-label extensions were included in this analysis: SELECT-EARLY (NCT02706873), SELECT-NEXT (NCT02675426), SELECT-MONOTHERAPY (NCT02706951), SELECT-COMPARE (NCT02629159), SELECT-BEYOND (NCT02706847), and SELECT-CHOICE (NCT03086343). Full details of the individual studies have been published previously and are summarized in Supplementary Table
S1 [
3‐
8].
Eligible patients were ≥ 18 years with active RA, defined as swollen joint count ≥ 6/66, tender joint count ≥ 6/68, and high-sensitivity C-reactive protein ≥ 3 mg/l (≥ 5 mg/l, together with evidence of erosive joint damage and/or autoantibody positivity, for SELECT-COMPARE and -EARLY). Of note, patients were required to have ALT and AST < 2 × upper limit of normal (ULN) at screening.
Patients were MTX-naïve (SELECT-EARLY) or had an inadequate response or intolerance to MTX (SELECT-MONOTHERAPY and -COMPARE), csDMARDs (SELECT-NEXT), or bDMARDs (SELECT-BEYOND and -CHOICE; Supplementary Table
S1). All clinical studies were conducted according to the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use guidelines, applicable regulations and guidelines governing clinical study conduct, and the Declaration of Helsinki. As this was an integrated analysis of pooled clinical trial data, IRB approval was not required.
Dosing
Per each study protocol, patients received extended-release upadacitinib (15 mg or 30 mg once daily [QD]), placebo, MTX, subcutaneous adalimumab (40 mg every other week [EOW]) or intravenous abatacept (weight-based dosing), as monotherapy or in combination with background csDMARDs. Patients were not permitted to switch between upadacitinib doses, although following a protocol amendment due to the approval of the upadacitinib 15-mg dose, patients receiving upadacitinib 30 mg were switched to the 15-mg dose (earliest switch at week 108 [SELECT-EARLY]/168 [SELECT-NEXT]/132 [SELECT-MONOTHERAPY]/180 [SELECT-BEYOND]). MTX-naive patients randomized to MTX started oral medication at 10 mg/week (7.5 mg/week in China and Japan) and were titrated to a maximum of 20 mg/week (15 mg/week in Japan) through week 8, according to tolerance (Supplementary Table
S1). Of note, patients receiving abatacept in SELECT-CHOICE switched to upadacitinib 15 mg QD at week 24, per the study protocol; thus, long-term data for abatacept are not available for this treatment group and were not included in this analysis.
The proportions of patients experiencing potentially clinically significant laboratory parameters at any time point were summarized for: pooled upadacitinib 15 mg QD, pooled upadacitinib 30 mg QD, adalimumab 40 mg EOW + MTX (SELECT-COMPARE only), and MTX monotherapy (SELECT-EARLY only). Patients who switched from placebo, adalimumab, or MTX to upadacitinib were included in the upadacitinib analysis set from the start of upadacitinib treatment. Those who switched from upadacitinib to adalimumab were included in the adalimumab analysis set from the start of adalimumab and were censored at time of switch. MTX monotherapy was censored at time of rescue to combination therapy (addition of a csDMARD). Serum samples for laboratory testing were collected from fasting patients where possible (minimum 8-h fast); if a patient was unable to fast when necessary, non-fasting status was recorded in the study source documentation.
Safety Analyses and Statistics
AEs and laboratory assessments were measured through August 15, 2022. AEs were defined using standardized Medical Dictionary for Regulatory Activities (MedDRA) query or company MedDRA query search criteria. A treatment-emergent adverse event (TEAE) was defined as any AE with an onset date from the first dose of study drug, up to 30 days after the last dose of either placebo or upadacitinib, or up to 70 days after the last dose of adalimumab (if patients discontinued from the study). Exposure-adjusted event rates (EAERs) were calculated as the total number of events (including multiple events in the same patient) adjusted for total exposure, reported as events per 100 patient-years (PY). Laboratory-related TEAEs of special interest were defined according to investigator-reported AEs (MedDRA query criteria as defined above) as opposed to objective laboratory measures. All events were attributed to treatment at the time of event.
Assessment of laboratory abnormalities included hemoglobin, neutrophils, lymphocytes, ALT, AST, CPK, creatinine, and lipids (low-density lipoprotein [LDL] and high-density lipoprotein [HDL] cholesterol, including the LDL/HDL cholesterol ratio [LDL-C/HDL-C]). Toxicity grading of laboratory-related AEs was based on Outcome Measures in Rheumatology (OMERACT) criteria [
22]. For grading of CPK and creatinine abnormalities, National Cancer Institute standard common terminology criteria grading methodology was used [
23]. To be included as grade 3 or 4 abnormality, the post-baseline grade must have been higher than baseline. Lipid data are reported as mean change from baseline. The studies were not designed to statistically compare the incidence of laboratory abnormalities and so all analyses are descriptive.
Ethical Approval
All clinical studies were conducted according to the International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use guidelines, applicable regulations and guidelines governing clinical study conduct, and the Declaration of Helsinki. As this was an integrated analysis of pooled clinical trial data, IRB approval was not required.
Discussion
This analysis assessed the impact of upadacitinib treatment on laboratory parameters in patients with RA in the context of clinical trials. Consistent with previous safety analyses [
24], we observed decreases in hemoglobin, neutrophils, and lymphocytes, and increases in liver enzymes and CPK in a proportion of patients receiving upadacitinib. There was no apparent association of these changes with the rates of serious or opportunistic infection, herpes zoster, or drug-induced liver injury. The incidence of laboratory abnormalities was generally higher with the upadacitinib 30-mg dose compared with upadacitinib 15 mg, supporting the choice of upadacitinib 15 mg as the approved dose for RA given the similar efficacy of the two doses [3–8]. Most patients who experienced abnormalities remained on upadacitinib.
Many treatments used in the management of RA are associated with abnormalities in laboratory parameters and require routine monitoring [
25]. MTX and other csDMARDs may result in liver enzyme and creatinine elevations and decreases in neutrophils. Indeed, this analysis reported elevations in liver enzymes in the MTX monotherapy group, although no creatinine elevations and few decreases in neutrophils were observed for this treatment group. The lack of effect on creatinine is consistent with previous results showing that low-dose MTX does not cause kidney-related AEs, in contrast to the high doses of MTX used in oncologic indications [
26]. In addition, this analysis reported higher rates of lymphopenia in the MTX-monotherapy group versus the adalimumab + MTX group. While we cannot provide one clear explanation, this could be a result of the smaller sample size (
N = 314 versus
N = 579, respectively) and PYs reported (865.1 PY versus 1573.2 PY, respectively). Given that AEs were reported at the investigator’s discretion, another explanation could be variability in how lymphopenia events were reported between study investigators. Among bDMARDs, interleukin-6 receptor inhibitors have been associated with neutropenia, thrombocytopenia, elevated liver enzymes, and lipid abnormalities [
27,
28]. In contrast, TNF inhibitors appear to have minimal clinically meaningful effects on laboratory parameters and monitoring is not typically recommended, although follow-up laboratory assessments often take place due to the concomitant use of csDMARDs [
25]. This is consistent with results from the current analysis, as changes in laboratory parameters observed in the adalimumab + MTX treatment group were generally small.
Across the JAK inhibitor class and within the upadacitinib data in the current analysis, increased serum levels of CPK in patients with inflammatory disorders have been observed. Although typically indicating muscle damage, in vitro data suggest JAK inhibition restores differentiation of myoblasts into mature myocytes and increases CPK expression [
29]. While there were two cases of rhabdomyolysis observed in the present study, these were resolved and were identified as not related to the study drug, per investigator. Although high CPK and creatine expression was reported in one of these cases, an alternative etiology to consider is the incidence of influenza and subsequent acute renal failure. Transaminase elevations have also been reported with JAK inhibitors for the treatment of RA, including tofacitinib and baricitinib [
15,
16]. Elevations have similarly been reported in patients treated with upadacitinib 15 mg and 30 mg, with most elevations not resulting in treatment discontinuation and being resolved or resolving regardless of upadacitinib discontinuation [
24]; results from our analysis are consistent with these findings.
It has been suggested that changes in laboratory parameters may vary according to JAK selectivity. For example, inhibition of JAK2 is thought to be related to hematologic abnormalities due to its involvement in erythropoietin and thrombopoietin signaling, whereas inhibition of JAK3 alters immune cell populations, due to its role in downstream signaling from common gamma chain cytokines [
30‐
33]. Therefore, the dose-dependent decreases in hemoglobin across upadacitinib treatment groups in the current analysis may be attributed to the inhibitory effect of upadacitinib on JAK2, consistent with results from ex vivo studies using peripheral blood mononuclear cells [
34]. Transient decreases in hemoglobin have also been observed in patients receiving baricitinib, which inhibits both JAK1 and JAK2 [
35]. However, further research is needed to elucidate the underlying mechanisms of the effect of upadacitinib on hemoglobin.
In addition, changes in circulating lipid levels have been observed with JAK inhibitors including upadacitinib [
13,
36]. The LDL-C/HDL-C ratio is considered a reliable predictor of cardiovascular mortality [
37] and remained consistent for upadacitinib 15 mg QD in the current analysis. Cardiovascular events are a relevant safety concern with JAK inhibitors, particularly due to the results of the ORAL Surveillance study, which found that tofacitinib was associated with an increased risk of major adverse cardiovascular events (MACE) compared with TNF inhibitors in patients with a history of atherosclerotic cardiovascular disease or other cardiovascular risk factor(s) [
38,
39]. In previous analyses of patients receiving upadacitinib, generally increased levels of LDL-C and HDL-C have been observed with no significant impact on cardiovascular disease risk [
36], and no apparent association has been found between LDL-C and occurrence of MACE [
24]. In this analysis, there was no evidence to suggest that elevated lipids were associated with the occurrence of treatment-emergent MACE. Further research may be needed to explore the association between changes in lipid levels and long-term cardiovascular events.
Overall, changes in laboratory parameters are a known effect of the JAK inhibitor class, and the results reported in this analysis are comparable to those observed in studies of baricitinib, tofacitinib, and filgotinib [
40‐
42]. The prescribing information for JAK inhibitors recommends assessing the levels of lymphocytes, neutrophils, hemoglobin, liver enzymes, and lipids at baseline and at regular intervals during treatment [
13,
15‐
18]. The prescribing information also recommends that treatment should not be initiated, or should be interrupted, in patients with a lymphocyte count of < 500 cells/mm
3, a neutrophil count of < 1000 cells/mm
3, hemoglobin levels of < 8 g/dl, or if liver injury is suspected, and that elevated lipid levels should be managed according to clinical guidelines for hyperlipidemia [
13,
15‐
18].
The main limitations of this study are the post hoc nature of the pooled analysis and the lack of a placebo control (although active comparators were included). There were also limited data available for adalimumab and MTX groups compared with upadacitinib. Finally, the studies included in this pooled analysis were not designed for direct or statistical cross-trial comparison between upadacitinib, adalimumab, and MTX, and differences in study design and patient characteristics should be considered and may limit interpretation of the data. The main strength of this study is the large patient population covering six phase 3 trials and > 10 000 PY of exposure to upadacitinib.
Declarations
Conflict of Interest
Christina Charles-Schoeman has received research grants from AbbVie, Alexion, Bristol-Myers Squibb, CSL Behring, Pfizer, and Priovant Therapeutics; and consulting fees from AbbVie, Bristol-Myers Squibb, Gilead, Pfizer, Recludix, and Sanofi/Regeneron. Jon T. Giles has received research grants from Pfizer; and consulting fees from AbbVie, Bristol-Myers Squibb, Eli Lilly, Gilead, Pfizer, and UCB. Nancy E. Lane has received consulting fees from Amgen, ANI, GSK, and Mallinckrodt. Ernest Choy has received research grants from Bio-Cancer, Biogen, Novartis, Pfizer, Roche, Sanofi, and UCB; consulting fees from AbbVie, Amgen, Biocon, Biogen, Chugai, Eli Lilly, Fresenius Kabi, Gilead, Janssen, Merck Serono, Novartis, Pfizer, Regeneron, Roche, R-Pharm, and Sanofi; and speaker fees from AbbVie, Amgen, Bristol-Myers Squibb, Chugai, Eli Lilly, Fresenius Kabi, Galapagos, Gilead, Janssen, Novartis, Pfizer, Regeneron, Roche, Sanofi, and UCB. Daniel E. Furst has received research grants from Actelion, Amgen, Bristol-Myers Squibb, Corbus, Galapagos, GSK, NIH, Novartis, Pfizer, Roche/Genentech, and Sanofi; consulting fees from AbbVie, Actelion, Amgen, Bristol-Myers Squibb, Corbus, Galapagos, Novartis, and Pfizer; and speaker fees from AbbVie, Continuing Medical Education, and Novartis. Jiří Vencovský has received consulting fees from AbbVie, Boehringer Ingelheim, Eli Lilly, Gilead, and Octapharma; and speaker bureau fees from AbbVie, Biogen, MSD, Pfizer, Roche, Sanofi, and UCB. Anthony G. Wilson has nothing to disclose. Gerd R. Burmester has received consulting fees and speaker bureau fees from AbbVie, Eli Lilly, Gilead, Janssen, MSD, Pfizer, Roche, and UCB. Derek Coombs, Sara K. Penn, Nasser Khan, Jillian B. Yee, and Kassim Rahawi are employees of AbbVie and may own stock or options. Iain B. McInnes has received research grants from Celgene, Eli Lilly, Janssen, Novartis, Pfizer, Roche, and UCB; and consulting fees from AbbVie, Amgen, Bristol-Myers Squibb, Cabaletta, Celgene, Compugen, Eli Lilly, Eyelo, Gilead, Janssen, Novartis, and UCB.